Task-Oriented Dialogue as Dataflow Synthesis

TL;DR

This paper introduces a novel dataflow graph-based approach to task-oriented dialogue, enhancing the representation and prediction of complex user intents through explicit metacomputation and a new dataset, SMCalFlow.

Contribution

It presents a new dataflow graph representation for dialogue state, a dataset with complex dialogues, and demonstrates improved model performance using this approach.

Findings

Dataflow graphs improve dialogue state representation.

Metacomputation operators aid in intent prediction.

Approach matches state-of-the-art on MultiWOZ.

Abstract

We describe an approach to task-oriented dialogue in which dialogue state is represented as a dataflow graph. A dialogue agent maps each user utterance to a program that extends this graph. Programs include metacomputation operators for reference and revision that reuse dataflow fragments from previous turns. Our graph-based state enables the expression and manipulation of complex user intents, and explicit metacomputation makes these intents easier for learned models to predict. We introduce a new dataset, SMCalFlow, featuring complex dialogues about events, weather, places, and people. Experiments show that dataflow graphs and metacomputation substantially improve representability and predictability in these natural dialogues. Additional experiments on the MultiWOZ dataset show that our dataflow representation enables an otherwise off-the-shelf sequence-to-sequence model to match the best existing task-specific state tracking model. The SMCalFlow dataset and code for replicating experiments are available at https://www.microsoft.com/en-us/research/project/dataflow-based-dialogue-semantic-machines.